T-lymphocyte function after retroviral-mediated thymidine kinase gene ...

© 2000 Nature America, Inc. 0929-1903/00/$15.00/⫹0 www.nature.com/cgt

T-lymphocyte function after retroviral-mediated thymidine kinase gene transfer and G418 selection Mauro Di Ianni,1 Sabrina Di Florio,1 Gigliola Venditti,1 Concetta Liberatore,2 Francesca Lucheroni,1 Franca Falzetti,1 Adelmo Terenzi,1 Carmelo Carlo Stella,3 Fabrizio Spinozzi,2 Patrice Mannoni,4 Massimo F. Martelli,1 and Antonio Tabilio1 Sections of 1Haematology and Clinical Immunology and 2Internal Medicine and Oncology, Department of Clinical and Experimental Medicine, Perugia University, Perugia, Italy; 3Bone Marrow Transplantation Unit, Istituto Nazionale dei Tumori, Milano, Italy; and 4Gene Therapy Center, Institut Paoli-Calmettes, Marseilles, France. Generation of an efficient graft-versus-leukemia (GVL) effect in patients with hematological malignancies who relapse after allogeneic bone marrow transplantation depends in part upon the number of infused T lymphocytes. Currently, a GVL reaction cannot be achieved without inducing concomitant graft-versus-host disease (GVHD); thus, one strategy is to try to modulate this GVL/GVHD ratio. We engineered human T lymphocytes with herpes simplex virus-thymidine kinase and neomycin resistance genes, with an LXSN-derived vector that confers a ganciclovir-specific sensitivity to the transduced T cells. We analyzed proliferation, interleukin-2 production, alloreactivity in a mixed lymphocyte culture, and clonogenicity during the different stages of retroviral infection and G418 selection. Our results confirm that a sufficient number of transduced T lymphocytes can be obtained after selection for clinical studies. Their proliferative activity, alloresponsiveness, and ability to produce and respond to interleukin-2 were retained. Compared with control populations, their clonogenicity, as assessed by limiting dilution assays, was reduced after retroviral infection and G418 selection by 1.6 and 2.9 logs, respectively, with both viral supernatant incubation and coculture procedures. This study shows that infection and selection with the thymidine kinase-neomycin resistance gene retroviral vector significantly reduces the number of functional T lymphocytes. This finding should be taken into account when establishing the dose of T lymphocytes necessary to trigger a modulated GVL/GVHD effect. Cancer Gene Therapy (2000) 7, 920 –926

Key words: Thymidine kinase; gene transfer; T lymphocytes; limiting dilution assay.

M

alignant hematological diseases are usually treated with induction chemotherapy, followed by consolidation therapy, the best of which is allogeneic bone marrow transplantation (BMT). Successful disease eradication depends upon the conditioning regimen and the graft-versusleukemia (GVL) reaction (i.e., the allogeneic suppression of residual neoplastic cells by donor anti-host cytolytic T cells).1 However, GVL is intimately associated with acute and chronic graft-versus-host disease (GVHD), which is the major cause of morbidity and mortality after allogeneic BMT.2,3 The occurrence of the GVL effect and GVHD both depend upon the number of donor T lymphocytes that are infused. The maximum dose of T lymphocytes to avoid GVHD is ⬍1 ⫻ 105 kg/body weight in matched allogeneic transplants4 and 1 log lower in mismatched transplants,5 but a much higher dose is required to develop Received April 2, 1999; accepted December 4, 1999. Address correspondence and reprint requests to Dr. Antonio Tabilio, Section of Haematology and Clinical Immunology, Department of Clinical and Experimental Medicine, Perugia University, Policlinico Monteluce, viale Brunamonti, 06122 Perugia, Italy. E-mail address: [email protected]

920

an effective GVL effect.6,7 Overcoming GVHD while maintaining an effective GVL effect is the main challenge in eradicating neoplastic clones. Gene engineering with suicide genes of normal and neoplastic cells makes target cells sensitive to a chemotherapeutic agent that is not normally toxic.8 The thymidine kinase (TK) gene of the herpes simplex type I virus (HSV-TK) is a suitable candidate because it selectively phosphorylates ganciclovir (GCV), and leads to its incorporation into DNA, causing cell death. The neomycin resistance (neoR) gene encodes for neomycin phosphotransferase II, an enzyme that protects the cells from the toxic effects of a neomycin analog, G418. Cells surviving exposure to G418 are assumed to express both the neoR and HSV-TK genes, with the HSV-TK gene expression resulting in a specific GCV-induced growth inhibition.9 –11 Consequently, these engineered lymphocytes could be used to trigger a controlled GVL/GVHD reaction, thus making repeated infusions of adequate doses of T lymphocytes feasible. Although it is possible to obtain a large number of HSV-TK/neoR-engineered T lymphocytes, very few data are available on their function after the transduction and selection processes.

Cancer Gene Therapy, Vol 7, No 6, 2000: pp 920 –926

DI IANNI, DI FLORIO, VENDITTI, ET AL: TK GENE TRANSFER IN T LYMPHOCYTES

921

added daily for 3 days. After transduction, cells were resuspended in fresh medium; after 24 hours, cells were cultivated with G418 (Life Technologies) at a concentration of 0.6 mg/mL for 7 days. A concentration of 0.3 mg/mL for 3 days was also tested. A lymphoid cell line, MOLT-4,14 was transduced and selected in the same way; the MOLT-4-1-SDF clone, which was G418-resistant and GCV-sensitive, was isolated. Figure 1. Vector design.

Proliferation assay In this study, we analyzed cell proliferation, the production of interleukin-2 (IL-2), alloreactivity, and the clonogenic potential of HSV-TK/neoR-transduced selected T lymphocytes. Our results confirm that T lymphocytes can be transduced on a large scale with the HSV-TK/neoR genes. However, retroviral vector infection and G418 selection significantly reduce T-lymphocyte clonogenicity as defined by limiting dilution analysis (LDA). Because this assay is a very sensitive method for detecting and quantifying functional T lymphocytes, our results could have important implications for defining the number of engineered T lymphocytes required to generate an efficient GVL effect. MATERIALS AND METHODS

Description of vector LSN-TK is a retroviral vector derived from the Moloney murine leukemia virus. This vector contains the HSV-TK gene controlled by the long terminal repeat promoter as well as the neoR gene transcribed from the early promoter of an internal simian virus 40. The HSV-TK gene was obtained by digestion of the MFG/ TK-IRES-LacZ vector12 with NcoI/BamHI restriction enzymes. The fragment was cloned in the pBlueBac4 NcoI site. pBlueBac4-HSV-TK was digested with EcoRI/BamHI, and the HSV-TK gene was obtained. The LXSN vector was digested with EcoRI/BamHI and HSV-TK gene was inserted at this site, thus obtaining the LSN-TK vector (Fig 1). The vector-producing cell line GP⫹envAM12/LSN-TK produced 1 ⫻ 107 colonyforming units/mL and was free of replication-competent viruses as assessed by mobilization assay.13

Collection, processing, transduction, and selection of donor lymphocytes We transduced lymphocytes on a large scale, with lymphocytapheresis products obtained from healthy donors. Mononuclear cells were separated on a Ficoll-Hypaque gradient and cultivated in RPMI 1640 medium (Life Technologies, Gaithersburg, Md) supplemented with 10% human AB serum (BioWhittaker, Walkersville, Md), penicillin/streptomycin, and L-glutamine. Monocytes were removed from the mononuclear fraction by adherence in 100-mm Petri dishes for 2 hours at 37°C. Lymphocytes in complete medium were then stimulated with phytohemagglutinin (2 ␮g/mL; Biochron KG, Berlin, Germany) and recombinant human IL-2 (100 U/mL; (Eurocetus B.V., Amsterdam, The Netherlands). After 3 days in culture, cells were transduced by cocultivation on a mitomycinpretreated layer in the presence of polybrene (4 ␮g/mL; Sigma, St. Louis, Mo) for 96 hours. Transduction was also performed with cells incubated with viral supernatant containing polybrene (4 ␮g/mL) 1:1 (vol/vol). Fresh viral supernatant was

Cancer Gene Therapy, Vol 7, No 6, 2000

The response of transduced lymphocytes and untransduced controls to IL-2 was detected and quantified by measuring, in a ␤ scintillation counter, the incorporation of tritiated thymidine ([3H]thymidine deoxyribose) after a 1-week cultivation with G418. Results are expressed as counts per minute and are presented as a mean of 10 different procedures ⫾ SD.

Clonogenic assay Clonable lymphocytes were assessed in populations of untransduced, transduced unselected, transduced selected, and transduced selected GCV-treated cell populations by LDAs.15 T cells were put into the clonogenicity assay after G418 selection ended. These clonogenicity experiments were repeated after a 5-day delay from the end of G418 selection. Samples from the four populations were seeded on the same day and analyzed after the same period of culture to limit the influence of aspecific culture conditions on clonogenicity. Briefly, irradiated peripheral blood mononuclear cells from normal, healthy subjects were used as feeder cells. Feeder and test lymphocytes are fully mismatched. A total of 10,000 –5,000 to 312 untransduced, transduced unselected, transduced selected, and transduced selected GCV-treated cells were plated in groups of 32 in 96-well, round-bottom microplates containing 1 ⫻ 105 irradiated feeder cells and brought to a final volume of 0.2 mL by adding RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum and IL-2 (100 U/mL). Plates were incubated at 37°C in a humidified atmosphere with 5% CO2. Microcultures were supplemented on day 6 or 7 with irradiated feeder cells and IL-2. Control wells contained irradiated feeder cells and IL-2 but no responder cells. Clonogenicity experiments were also performed on T lymphocytes transduced with MFG/nls-LacZ vector16 as an additional control. After 12–22 days of culture, each microwell was assessed microscopically for cell growth. The frequencies of proliferating cells were calculated by ␹2 test from the Poisson distribution relationship between the number of plated cells per culture and the logarithm of the percentage of nonproliferating cultures.

GCV killing assay Transduced selected lymphocytes and both transduced unselected and untransduced cells were cultured in the presence of GCV (1 ␮g/mL; Syntex, Palo Alto, Calif) for 10 days. The effects of GCV on cell growth are expressed as a percentage of viable cells (trypan blue-negative).

Allostimulation Transduced selected cells were washed and cultured for 5 days in 96-well, round-bottom plates (105 cells/well), together with IL-2 (100 U/mL), irradiated (4000 rad) allogeneic mononuclear cells (105 cells/well), and either with or without GCV (1 ␮g/mL). Radiolabeled thymidine ([3H]thymidine deoxyribose; 1 ␮Ci/well) was added for the last 12–16 hours culture before harvesting the DNA and counting in a ␤ scintillation counter.

922


Fluorescence-activated cell sorting analysis The phenotype was determined with a panel of mouse monoclonal antibodies directed against the following antigens: CD3, CD4, CD8, CD11, CD16, CD19, and CD25 (Coulter Corporation, Hialeah, Fla); HLA-DR (Ylem, Rome, Italy); and CD20 and CD56 (Immunotech, Marseilles, France) with an indirect fluorescence-labeling method. Cells were analyzed with an Epics XL cytometer (Coulter) on days 0, 7 (postinfection), and 14 (postselection). Intracellular IL-2 staining was performed with a directly conjugated anti-IL-2 antibody (Delta Biologicals, Rome, Italy) in conjunction with cell permeabilization techniques.17 Phenotyping and IL-2 staining were performed in populations of untransduced, transduced unselected, and transduced selected cells and on the clones obtained from LDA.

Polymerase chain reaction (PCR) The presence of neoR gene in untransduced, transduced unselected, and transduced selected lymphocytes was detected by PCR analysis. Lymphocytes (1 ⫻ 106) were lysed in 500 ␮L of a buffer containing 10 mM tris(hydroxymethyl)aminomethaneHCl, 50 mM KCl, 2.5 mM MgCl2, 0.1% gelatin, 0.45% Nonidet P-40, 0.45% Tween 20, and 100 mg/mL proteinase K. A total of 10 ␮L of extract was used for a PCR with the following neoR primers: 5⬘-GGTGGAGAGGCTATTCGGCTATGA-3⬘ and 5⬘TCCTGATCGACAAGACCGGCTTCC-3⬘, as reported previously.10 ␤-Actin DNA was amplified with the same protocol and was used as internal control for DNA integrity and loading. DNA was amplified under standard reaction conditions with 35 cycles of incubation at 94°C (for 30 seconds), 55°C (for 30 seconds), and 72°C (for 45 seconds), followed by a 10-minute extension period at 72°C. The amplified fragments were separated by electrophoresis on 2% agarose gel and stained by ethidium bromide. In each experiment, neoR amplification band intensity was compared with the calibration amplifications obtained from diluted TK-neoRtransduced MOLT-4 cells. PCR was also performed with [␣-32P]deoxycytidine triphosphate (1.5 ␮Ci/each tube) under the same experimental conditions described above. The PCR product was purified and eluted in Sephadex G50 Quick Spin columns (Roche Diagnostic, Basel, Switzerland). The final elution product was counted in a ␤ counter. Results are expressed as counts per minute. The percentage of infected cells was calculated by comparing the counts per minute for each sample with the counts per minute for the known dilution of TK-neoR-transduced MOLT-4 cells.

RESULTS

Efficiency of transduction A total of 10 transduction procedures were performed. A mean of 7 ⫻ 108 lymphocytes was transduced with viral supernatant, and 4 ⫻ 108 cells by cocultivation. During the transduction procedures, the number of cells doubled when supernatant was used and remained unchanged with cocultivation. After a 7-day selection with G418, the number of live cells remained unchanged because of the expansion of transduced cells and the death of untransduced cells. Cell samples before and after G418 selection were analyzed by semiquantitative PCR with both neoR and ␤-actin primers. The percentage of infected cells in-

Figure 2. a: Semiquantitative PCR was performed with a transduced selected lymphoid cell line (MOLT-4). DNA from transduced MOLT-4 cells was mixed in different proportions with untransduced cell DNA to generate a standard curve. Sample A is untransduced primary T lymphocytes, sample B is the HSV-TK-transduced unselected population with 1–5% transduced cells obtained with supernatant infection, sample C is the HSV-TK-transduced unselected population with 25– 30% transduced cells obtained with cocultivation, sample D is the HSV-TK- and G418-selected primary T lymphocytes showing ⬎90% transduction after supernatant infection, and sample E is the HSV-TKand G418-selected primary T lymphocytes with ⬎90% transduction after the cocultivation procedure. PCR analysis was repeated in each different lymphocytapheresis product used and gave overlapping results. b: Radioactive PCR was performed with a transduced selected lymphoid cell line (MOLT-4). DNA from transduced MOLT-4 cells was mixed in different proportions with untransduced cell DNA to generate a standard curve expressed in counts per minute. Sample A is untransduced primary T lymphocytes, sample B is the HSV-TKtransduced unselected population with 1–5% transduced cells obtained with supernatant infection, sample C is the HSV-TK-transduced unselected population with 25–30% transduced cells obtained with cocultivation, sample D is the HSV-TK- and G418-selected primary T lymphocytes showing ⬎90% transduction after supernatant infection, and sample E is the HSV-TK- and G418-selected primary T lymphocytes with ⬎90% transduction after the cocultivation procedure.

creased from 1–5% before selection to ⬎90% after G418 selection when viral supernatant infection was used and increased from 25–30% to ⬎90% in the cocultivation experiments (Fig 2, a and b). Figures 3-5 illustrate the results of lymphocyte transduction with viral supernatant. Cocultivation results overlapped with these results. The results presented in Figure 3 show how a 7-day selection with 0.6 mg/mL G418 resulted in a cell population that was inhibited by GCV treatment. In fact, a 10-day treatment with 1 ␮g/mL GCV eliminated the transduced selected population but had no significant effect on untransduced or transduced unselected populations, thus confirming that before G418 selection, the percentage of cells expressing the HSV-TK gene was very low. Interestingly, ⬎80% of TK-neoR-transduced cells were killed at this GCV concentration at day 10. No viable cells were observed after 14 days of culture. The transduced selected population, although less responsive to IL-2 than control cells, was still able to



923

Figure 3. Untransduced, transduced unselected, and transduced selected lymphocytes were cultured with GCV (1 ␮g/mL) for 10 days. More than 80% of HSV-TK-transduced cells were killed at this GCV concentration at day 10. No viable cells were observed after 14 days of culture. The figure shows the mean ⫾ SD of 10 different experiments using cells transduced with viral supernatant.

proliferate (Fig 4). After 1 week of G418 selection, some untransduced cells remained viable but lost their capability to respond to IL-2 and consequently their proliferative potential. A 3-day selection with G418 was not sufficient to select a transduced population. In fact, untransduced populations were still viable at 15 days after the end of treatment (data not shown).

Functional activity of transduced cells Alloreactivity, as measured by the one-way mixed lymphocyte reaction, showed that G418 selection does not alter the alloresponsiveness of these cells. In fact, transduced selected cells were still able to proliferate in presence of allogeneic stimuli. Furthermore, a 5-day GCV treatment inhibited the growth of trans-

Figure 4. Response of untransduced and transduced cells to IL-2 (100 U/mL) at day 0 (left) and after 7 days of culture with G418 (0.6 mg/mL) (right). The response to IL-2 is suppressed but still present in transduced cells after G418 selection. The response to IL-2 is eliminated in untransduced cells after G418 selection. Counts per minute have been normalized to the number of cells. The figure shows the mean ⫾ SD of 10 different experiments using cells transduced with viral supernatant.


924


Figure 5. GCV-induced growth inhibition of alloreactive transduced selected T cells. Transduced selected cells were cultured with irradiated allogeneic cells and IL-2 (100 U/mL), either with (䡺) or without (F) GCV (1 ␮g/mL) for 5 days.

duced selected cells compared with untreated cells (Fig 5). Cloning efficiency was 0.8 per 1 plated cell in controls (untransduced lymphocytes). LDA indicated that the number of clonable T lymphocytes was 1.6 logs less in the transduced unselected population and 2.9 logs less in transduced selected cells (Fig 6). No differences were found between cells transduced by cocultivation or with viral supernatant. When clonable lymphocytes in the transduced unselected and transduced selected populations were compared, a difference of 1.3 logs was found. Both transduction and selection procedures strongly reduce clonable lymphocytes (Fig 6). No clonable lymphocytes were found after GCV treatment in the transduced selected population. When these clonogenicity experiments were repeated 5 days after the end of G418 selection, another 1 log decrease in clonable T cells was seen. When transduction was performed with nls-LacZ as an irrelevant gene, no difference was found in clonable cells compared with the TK-neoR-transduced unselected population. HSV-TK transduction and G418 selection did not change the lymphocyte phenotype. Trivial differences emerged between controls and infected populations at all timepoints, demonstrating that retroviral infection has minimal effects on lymphocyte subset composition (Fig 7).

Intracellular IL-2 staining showed a percentage of positive cells ranging from 42.8 to 50% in untransduced cells, 41.5 to 48% in transduced unselected cells, and 16.08 to 22.34% in transduced selected lymphocytes. The mean IL-2 production in clones of the transduced selected population was 34.8%. DISCUSSION Donor T lymphocytes play a crucial role in allogeneic BMT because they facilitate engraftment, help eradicate residual neoplastic cells, and promote resistance to and/or overcome potentially lethal virus-induced diseases such as cytomegalovirus pneumonia. Finally, they are used to treat Epstein-Barr virus-induced lymphoproliferative disease and hematological malignancies that relapse after BMT.6,7,18–20 Their anti-leukemic effect might be mediated by major histocompatibility complex-unrestricted natural killer cells, by lymphokine-activated killer cells, or by major histocompatibility complexrestricted T lymphocytes.1 Although some findings suggest that lower doses of T cells or CD8-depleted donor lymphocytes may mediate a GVL effect without clinical GVHD,19–21 GVL is intimately associated with GVHD, and the number of T lymphocytes determines the incidence and severity of

Figure 6. Effects of HSV-TK transduction and G418 selection on T-lymphocyte clonogenicity. The figure shows the mean ⫾ SD of 10 different experiments.



925

Figure 7. Immunophenotypic analysis of untransduced, transduced, and selected lymphocytes.

GVHD. Indeed, as few as 1 ⫻ 105 cells/kg clonable T cells cause GVHD in matched allogeneic BMT, if given on the day of transplant4 and the dose is 1 log lower in mismatched BMT.5 GVHD in patients receiving adoptive immunotherapy for relapses after BMT was apparently initiated by higher numbers of alloreactive donor T cells (i.e., ⬎1 ⫻ 108 cells/kg as previously demonstrated by experimental22 and clinical data).19,21 Most studies show that a significant clinical response is obtained with T-lymphocyte doses of ⬎1 ⫻ 108 cells/kg.7,19,21 Consequently, complete remissions by means of donor lymphocyte infusions are offset by the risk of GVHD, which occurs in 60–90% of responders with a reported mortality of 20%.20 Several different strategies have been proposed in an attempt to separate GVL from GVHD. The use of lower doses of T cells or CD8-depleted T cells seems to be associated with a low rate of severe acute GVHD.19 –21 An innovative alternative is generating a controlled GVL/GVHD by means of T lymphocytes engineered with suicide genes.23 The feasibility of transducing T lymphocytes with a retroviral vector containing the HSV-TK gene has been reported.10,11 Although T lymphocytes maintain their biological functions (i.e., IL-2 production and alloreactivity after TK transduction),10,11 their clonogenic capacity has not been investigated to date. This report confirms that an efficient HSV-TK-neoR transduction, resulting in a specific GCV growth inhibition of primary T lymphocytes, is possible after G418 selection, and that a large purified population of transduced cells is made available for in vivo studies. Although at a lower rate than control cells, transduced cells retained their capacity to produce and respond to IL-2 and their alloresponsive-


ness in one-way mixed lymphocyte culture, all of which indicate their functional integrity. However, clonogenic analysis of T lymphocytes after retroviral vector infection and G418 selection shows a 1.6 log10 and 2.9 log10 respective loss of clonogenicity as defined by LDA. This assay is extremely sensitive in quantifying functional T lymphocytes4 and provides an accurate estimate of the number of alloreactive T lymphocytes that survive the entire engineering procedure. This result is confirmed by the finding that IL-2 production in the transduced selected clones was higher than in the total transduced selected population, suggesting that not all transduced selected cells are capable of proliferating in a medium-/long-term period. Shortening the culture time to 3 days, as already reported,11 does not ensure, under our experimental conditions, an optimum selection as shown by the survival of untransduced T cells after GCV treatment. The data from this study show that both the gene transfer process (by supernatant incubation and cocultivation) and G418 selection always reduce the clonogenic capacity of T lymphocytes. Hence, our results are particularly noteworthy because most ongoing clinical trials are based on retroviral infections and/or on G418 selection of TK-transduced T lymphocytes.11,24,25 Current clinical protocols aim to assess the effects of gene-modified T cells on bone marrow engraftment and tumor relapse or to evaluate the efficacy of therapy for Epstein-Barr virus-induced lymphoproliferative disease or leukemias relapsing after allogeneic BMT.11,24,26 These protocols trigger controlled GVL/GVHD by administering escalating doses of TK-transduced lymphocytes starting from 1 to 2 ⫻ 105 cells/kg. Consequently, assuming the

926


infection-selection procedure reduces T-lymphocyte clonogenicity by ⬃3 logs, the total number of alloreactive T lymphocytes that are infused should be taken into account to maximize the efficacy of this therapy. In conclusion, this study provides critical information on the functions of HSV-TK-transduced T lymphocytes. We confirm that these lymphocytes are still biologically active, but the number of alloreactive T cells is significantly lower than expected. Because LDA quantifies the number of clonable lymphocytes capable of responding to allostimulation, we suggest using this sensitive, specific, and reproducible test to evaluate the number of HSV-TK-transduced T cells that are really capable of triggering an efficacious GVL/GVHD reaction. Finally, functional, precise quantification of transduced T lymphocytes is mandatory to establish the threshold dose of GVL/GVHD, especially in T-cell-depleted, unrelated, or mismatched transplants where GVHD is serious and difficult to control.5,27 ACKNOWLEDGMENTS We thank Dr. Geraldine Anne Boyd for her helpful criticism and comments and Regina Lupi for her secretarial help. This study was supported by a grant from the Fondazione Cassa di Risparmio (Perugia, Italy).

REFERENCES 1. Barrett AJ. Mechanisms of the graft-versus-leukemia reaction. Stem Cells. 1997;15:248 –258. 2. Gale RP. Graft-versus-host disease. Immunol Rev. 1985; 88:193–214. 3. Ferrara JL, Deeg HJ. Graft versus host disease. N Engl J Med. 1991;324:667– 674. 4. Kernan NA, Collins NH, Juliano L, Cartagena T, Dupont B, O’Reilly RJ. Clonable T lymphocytes in T-cell-depleted bone marrow transplants correlate with development of graft-v-host disease. Blood. 1986;68:770 –773. 5. Reisner Y, Martelli MF. Bone marrow transplantation across HLA barriers by increasing the number of transplanted cells. Immunol Today. 1995;16:437– 440. 6. Porter DL, Roth MS, McGarigle C, Ferrara JLM, Antin JH. Induction of graft-versus-host disease as immunotherapy for relapsed chronic myeloid leukemia. N Engl J Med. 1994;330:100 –106. 7. Lokhorst HM, Schattenberg A, Cornelissen JJ, Thomas LLM, Verdonck LF. Donor leukocyte infusions are effective in relapsed multiple myeloma after allogeneic bone marrow transplantation. Blood. 1997;90:4206 – 4211. 8. Culver KW, Raw Z, Wallbridge S, Ishii H, Oldfield EH, Blaese RM. In vivo gene transfer with retroviral vectorproducer cells for treatment of experimental brain tumors. Science. 1992;256:1550 –1552. 9. Moolten FL. Tumor chemosensitivity conferred by inserted herpes thymidine kinase genes: paradigm for a prospective cancer control strategy. Cancer Res. 1986;46: 5276 –5281. 10. Tiberghien P, Reynolds CW, Keller J, et al. Ganciclovir treatment of herpes simplex thymidine kinase-transduced T lymphocytes: an approach for specific in vivo donor T-cell depletion after bone marrow transplantation? Blood. 1994;84:1333–1341.

11. Munshi NC, Govindarajan R, Drake R, et al. Thymidine kinase (TK) gene-transduced human lymphocytes can be highly purified, remain fully functional, and are killed efficiently with ganciclovir. Blood. 1997;89:1334 –1340. 12. Di Ianni M, Casciari C, Ciurnelli R, et al. Retroviral transfer of herpes simplex virus-thymidine kinase and ␤-galactosidase genes into U937 cells with bicistronic vector. Leukemia Res. 1997;21:951–959. 13. Varmus H. Retroviruses. Science. 1988;240:1427–1435. 14. Srivastava S, Minowada J. Terminal deoxynucleotidyl transferase activity in a cell line (molt-4) derived from peripheral blood of a patient with acute lymphoblastic leukemia. Biochem Biophys Res Commun. 1973;51:529 –535. 15. Chen WF, Wilson A, Scollay R, Shortman K. Limit dilution assay and clonal expansion of all T cells capable of proliferation. J Immunol Methods. 1982;52:307–322. 16. Humeau L, Chabannon C, Firpo MT, et al. Successful reconstitution of human hematopoiesis in the SCID-hu mouse by genetically modified, highly enriched progenitors isolated from fetal liver. Blood. 1997;90:3496 –3506. 17. North ME, Ivory K, Funauchi M, Webster AD, Lane AC, Farrant J. Intracellular cytokine production by human CD4⫹ and CD8⫹ T cells from normal and immunodeficient donors using directly conjugated anti-cytokine antibodies and three-colour flow cytometry. Clin Exp Immunol. 1996;105:517–522. 18. Kolb HJ, Schattenberg A, Goldman JM, et al. Graftversus-leukemia effect of donor lymphocyte transfusions in marrow-grafted patients: European Group for Blood and Marrow Transplantation Working Party Chronic Leukemia. Blood. 1995;86:2041–2050. 19. MacKinnon S, Papadopoulos EB, Carabasi MH, et al. Adoptive immunotherapy evaluating escalating doses of donor leukocytes for relapse of chronic myeloid leukemia after bone marrow transplantation: separation of graftversus-leukemia responses from graft-versus-host disease. Blood. 1995;86:1261–1268. 20. Van Rhee F, Kolb HJ. Donor leukocyte transfusions for leukemic relapse. Curr Opin Hematol. 1995;2:423– 430. 21. Alyea EP, Soiffer RJ, Canning C, et al. Toxicity and efficacy of defined doses of CD4⫹ donor lymphocytes for treatment of relapse after allogeneic bone marrow transplant. Blood. 1998;91:3671–3680. 22. Van Bekkum DW, Kinwel-Bohre´ EPM. Risk of donor lymphocyte infusions following allogeneic bone marrow transplantation. Exp Hematol. 1997;25:478 – 480. 23. Bonini C, Ferrari G, Verzelletti S, et al. HSV-TK gene transfer into donor lymphocytes for control of allogeneic graft-versus-leukemia. Science. 1997;276:1719 –1724. 24. Tiberghien P, Cahn JY, Brion A, et al. Use of donor T-lymphocytes expressing herpes-simplex thymidine kinase in allogeneic bone marrow transplantation: a phase I-II study. Hum Gene Ther. 1997;8:615– 624. 25. Fellowes WS, Bunnell BA, Carter CS, et al. Preparation of gene-transduced CD4 syngeneic lymphocytes for clinical antiHIV therapeutic trial. J Hematother. 1998;7:271. Abstract. 26. Link CJ, Burr RK, Traynor AE, Drobyski WR. Adoptive immunotherapy for leukemia: donor lymphocytes transduced with the herpes simplex thymidine kinase gene for remission induction. Hum Gene Ther. 1998;9:115–134. 27. Aversa F, Tabilio A, Terenzi A, et al. Successful engraftment of T-cell-depleted haploidentical “three-loci” incompatible transplants in leukemia patients by addition of recombinant human granulocyte colony-stimulating factor-mobilized peripheral blood progenitors cells to bone marrow inoculum. Blood. 1994;84:3948 –3955.
